Rigid image registration using parallel processing

Recently image registration has become a major area to work in image processing domain. Image registration is mainly based on transformation theory that refers to transforming one matrix or image with reference to other. Image processing has revolutionized medical image processing. With the help of image registration, different medical images captured by different medical imaging devices are registered, to analyze and post-process the images further. Although the main problem of image registration is its high time complexity. The objective of this paper is to solve the time complexity issue by using multithreading parallel processing technique. Apart from fastening the image processing, this work also analyzed the effects of parallel processing on image registration framework to further discuss the need of parallel processing in this domain. Several statistical methods such as Structural similarity Index, correlation has been used to analyze the registered images along with visualization of registered frames.